It’s sad to see that research in OS is actually dead. There’s no reason to do any more fancy things because the payback is not worth it because no sooner do you announce something new, you get a bunch of 12year olds screaming is it GPL’ed?.
To tell the truth, I don’t think this is all that recent of a development. With some exceptions, academic CS has diverged quite markedly from commercial CS. In particular, precious little of the academic OS research done in the last 15 years has found it’s way into production systems, while very few academic projects have been focused on the sort of problems currently at the forefront of commercial concerns.
In most engineering fields, there is a sort of cycle wherein academic research (including research-oriented divisons of commercial companies) leads commercial implementation by 5-10 years. However, that doesn’t seem to be the case in the computer industry. The current state-of-the-art in the commercial industry is a good 20 years behind academia. I think the inherent inertia of the computer industry may have a lot to do with this.
For example:
– When Lockheed makes an advancement in engine design, the cost of bringing that advancement to market is bounded by the cost of bringing it to production-quality.
– When IBM makes an advancement in computers, the cost of bringing it to market is unbounded. Not only does it have to be brought to a production level of quality, but they have to spend a tremendous amount of money in clearing out the existing product in that segment. This cost can dwarf the cost of actually developing the new product.
The bottom-line is that it’s simply not profitable for research CS and commercial CS to coexist any longer. The research folks have no interest in rehashing existing products, and the commercial folks cannot afford to shake things up too much. Thus, each goes their own way, serving their own needs.
if a bunch of 12year olds screaming is it GPL’ed? is the only thing thats holding you back from doing OS research then i guess you never was really thinking about doing it.
that would be like saying nope im not going to school cus some kids make noise during class i think il just stay home and sleep.
if you cant deal with such small issues then you have a problem and serisule need to deal with it
It’s interesting to note that there is a key component of the academia/commercial cycle that is drastically scaled-down in the computer industry. Commercial used to run highly research-oriented divisions dedicated to developing new, risky, potentially profitable ideas. These groups were key in mediating between sources of new ideas, and practical implementations of them.
These days, most of those have been drastically cut back. Apple’s Cambridge labs are closed. DEC is out of business. A lot of small, innovative companies have gone out of business. Xerox’s PARC and HP Labs have been doing a fraction of the work they used to. Microsoft spends a lot on research, but none of the cool stuff ever seems to show up in it’s products. IBM and Intel still do some good work, but they can’t carry the industry by themselves.
“if a bunch of 12year olds screaming is it GPL’ed? is the only thing thats holding you back from doing OS research then i guess you never was really thinking about doing it. ”
LOL! SO TRUE MAN! It’s so plain annoying screaming is it GPL’d? if its not GPL’d it will fail. You can’t beat the GPL. Mother, are you GPL’d? only good mothers and fathers are GPL’d.
There is a good reason why people ask “is it GPL.” The market has shown that small innovative companies have a very marginal chance of surviving. GPL (or some OSS license) gives a upstart technology the chance to survive independent of it’s company. As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
There is a good reason why people ask “is it GPL.” The market has shown that small innovative companies have a very marginal chance of surviving. GPL (or some OSS license) gives a upstart technology the chance to survive independent of it’s company. As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
Maybe it’s better to ask if it’s open-sourced instead of just GPL’d. The problem with BeOS is that none of the code was ever released after Palm bought it – except for some leaked bits I guess. If you read why Haiku is under a MIT/X11 license(I think), it’s because a major contributor gave a significant chunk of code to the project under the stipulation that it not be GPL’d. The GPL is just too viral for many people.
If you read why Haiku is under a MIT/X11 license(I think), it’s because a major contributor gave a significant chunk of code to the project under the stipulation that it not be GPL’d.
@Lumbergh: OSS, sure. I’m not pedantic. As long as the code doesn’t die with the company, I’m happy.
@Wolf: I’d appreciate it if you could post the research that led you to that conclusion. I’ve tried Googling for it, but all I turn up is uninformed speculation…
What is the income of all the linux vendors combined together? So how many employees can they hire? OSS specially GPL’ed software is killing UNIX business, SUN fired so many employees…did anyone hire the people who lost jobs at SUN?
GPL is now harming windows too and thousands of other companies make money off their products on windows…but they can’t do same on linux because linux folks are used to getting stuff for free. So in a way GPL is affecting the whole software industry ECOSYSTEM.
In a healthy ecosystem, universities will do research, industry will use their research and hire those research professional, market the product, make money off it…donate some money back to universities. It was more like a self feeding system.
With GPL in picture, the whole ecosystem is broken, somebody do research, marks their software GPL, companies now can’t hire them, unless they are thinking of making money off support (which you know aint much, because support is not a specialization), so due to GPL, industry is not benefited, these researcher can’t get a job, their product is not market’ed and thus total collapse of ecosystem. GPL is like a self eating and feeding system. Its Viral, infectious.
How much innovation is done in Linux world as compared to other commerical companies like SUN, Microsoft and IBM. Sure IBM can use OSS to make money off hardware but not every other company can do it.
Hence…i believe GPL is killing software industry. Richard Stallman is Hitler.
it is one possibility that GPLing, or the demand that you do is killing the OS and software market. Another possibility is that we are consuming all our coding energys in protecting our cash flow stream. If it were all GPLed then we would get over figuring out how keep our cash flow from Office or our OS or what ever. How much energy is wasted because one thing is not open source, so we have to go and make another thing just like it to make it work.
If we got over the idea of making our billons off of something here already today… our energy today is consumed damming up the stream for our interests. We should be widening opening up the stream so it flows fast and freely for all,,, sorta like the right to read.,,, that would free up the energy that is needed to make the next great project.
What you consider an ecosystem, I consider an outmoded hierarchy. People are free, and thus they are free to code as they please, without having to adhere to some feudalistic food chain. Any argument against the GPL is silly at face-value, because they argue that people acting freely and magnanimously is somehow a bad thing. In a free society, such arguments are unacceptable.
Open source came about because some people wanted things that commercial companies were not willing to provide. So they built these things themselves. Now, if commercial companies feel cheated, well, they dug their own graves. If the market is using Linux instead of buying your products, well, maybe you should think about making your product more attractive to the market. And if you cannot make your product more appealing (free is hard to argue with), then you simply just accept that the people do not want what you have to sell. Either way, the customer wins — he get’s a better product, or a cheaper product. And in a free-market society, what’s good for the customer is ultimately good for everyone.
In a healthy ecosystem, universities will do research, industry will use their research
One of the prime problems is that the industry is *not* using their research. Longhorn will mark Microsoft’s transition to peddling 1980’s technology, a step-up from the 1970’s technology it was peddling before. If industry had been following academia, well, then maybe some people would want to buy their products. For example: I’ve been musing about the shortcomings of IDEs (after heavily using KDevelop and VS.NET for the last few months). If IBM hadn’t given up it’s research on Smalltalk, and had spent the last 15 years working on a killer IDE for that, they’d have a product I’d shell out $2000 for in a blink. Instead, they gave that up, and are just now (with Eclipse) catching up to where they were in the early 1990’s.
Hence…i believe GPL is killing software industry. Richard Stallman is Hitler.
He is so correct! After all, since OSS (or is it FLOSS) is based on communistic ideals, then Stallman is Hitler! All of this ties into the terrorists that run Linux.
How dare people demand that the industry remain dynamic and mutate as the world changes? I think we should all ban cars as they killed the train industry (on of the most powerful and monopolistic industries back in the day)
Also all those fancy assembly lines caused all those poor workers to get fired. We want manual labor back!
And how the internet has practically transformed enourmus areas of the economy (banking, ebay, etc…).
Let as not embrace advancements if it should, heaven forbid, force the economy to remain dynamic and change to fit the needs of the time.
</sarcasm>
Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer), so the company gets considerably more then its money back per copy. In the software economy what costs money to the company is not something that is per product, like say in the car/most other industries. Of course in the car industry you are paying for the reaserch and design that went into the car as it is piggybacked onto the price of the car, but, compared to the price of the actual car, it is not that big. Simmilar things exsist in the Music/Video/Book(?) industry, but there the piggybacking of the non-per-copy cost is also relativly small.
A bigger problem is the fact that once a user buys an OS, he is locked into the OS and possibly the hardware. This is something that is practically unique to the software industry.
As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
This is a myth. How many platforms have actually died?
If a system has users it usually continues in some form often with multiple choices:
For Amiga users there are 2 boards and 3 operating systems to choose from.
The Sinclair QL “died” in the early 80’s but you can get variations today.
You can get RiscOS machines from multiple manufacturers.
If you want BeOS you can still get it, but it’s called Zeta now. Haiku will take a while yet but it’ll be available sometime and that’s only 2 of the choices.
What does usually seem to happen is a pause, sometimes lasting years but many platforms have continued and ended up vastly more advanced than the original. If anything the “death” of a system may be one of the best things that can happen to it!
Geniuses like Wolf here can be naturally-born, I guess, but they’re often products of machinery that has interests in their unusual intelligence and is also willing to support it financially.
I’m here mostly for the comments and I sure wouldn’t be that democratic, if it was my site.
He is so correct! After all, since OSS (or is it FLOSS) is based on communistic ideals, then Stallman is Hitler!
I’m going to post even though Godwin’s Law has been called. 😉 Just a history correction. Nazi’s were strongly anti-communist. If OSS is based on communistic ideals, Stallman must be Joseph Stalin.
The existance of FLOSS is not the consequence of Stallman’s ideas, it is related to the very nature of software which can be copied at zero cost. There is no alternative. Software can be compared with Music and Litterature whose artworks are not based on physical artefacts ( unlike painting or theatre … )
CS research can be made at a very low cost, by individuals or in third world countries, it is basically a cheap science, comparing with Physics, Chemistry, Biology, … ( It is nearly as cheap as Maths ). Software patents is a mean to transform zero investment ideas into unfair concurrency barriers.
Commercial software vendors have abandoned for long the idea of catching up with CS research. I think that they fear progress. For example, in Lisp Machine or Smalltalk style environments, where every single bit of applications is available for reuse or tuning, it is impossible to sell something like a closed source word processor. Microsoft, Adobe, and others cannot follow CS research because really advanced computer environments are a threat for their existance ( Imagine an oil company doing research for propelling cars with water instead of petroleum ).
Why FLOSS developpers doesn’t try to make these ideas widely available still puzzles me.
( BTW, really advanced computers doesn’t need programmers, they will do programming themselves 🙂 <grin/>
( About Hitler & Communism : He was strongly anti-everyone. He arranged an anti-aggression pact with Stalin at the begginning of WW2, while prosecuting german communists. Communism is not about making everything free to everyone, it’s about nationalisation of production means, as an opposition to capitalism which encourages “the enrichment of a few at the expense of the working classes”. In other words, a communist US gov. would nationalize Microsoft and sell Windows $2 to everyone. IMHO, Stallman is more an Anarchist than a Communist. )
As I once said in osnews (and IIRC was promptly modded), the danger I see in having software (or anything) for free (or perceived as such) is stifle in innovation. Tthis has nothing to do with MS pr – MS isn’t innovative by any means, and the fact that its OS is ‘perceived as free’ – in the sense that when Joe user buys a computer it comes with it – is the greatest contributor to the mentioned stifling effect.
That happens because, when there is a tool for a job that you can have for nothing, you tend to stick with it, no matter its shortcomings. Just as the availability of slave labour discourages technological research (but especially adoption), so does the availability of ‘reasonable’ software.
In a market where you have to pay for products, you tend to pick what suits you best. In a monopoly (as MS) or a market where products are considered to come for free (the GPL), you are at a loss when it comes to decide what’s best. Both stifle competition. Now the latter is much preferable to the former as the individual user is concerned, but in large scale I’m not convinced the difference is big.
It would be nice if the ‘software industry’ were like most others: companies would ask a _fair_ price for their software, be open to customer requests, and struggle to make their products better! That is, the opposite of what went on in the 80s and 90s mainstream.
I’m going to post even though Godwin’s Law has been called. 😉 Just a history correction. Nazi’s were strongly anti-communist. If OSS is based on communistic ideals, Stallman must be Joseph Stalin.
Interesting enough, I once had an US right winger argue that nationalsocialism was in fact a “leftist” thing, much like communism. The proof? Well, it started with the name. Nationalsocialism. Some people…
GPL is now harming windows too and thousands of other companies make money off their products on windows…but they can’t do same on linux because linux folks are used to getting stuff for free. So in a way GPL is affecting the whole software industry ECOSYSTEM.
History has already proven you wrong. The BSD people (original BSD Unix) gave away their os on tape with all source. This didn’t stop companies hiring those BSD hackers (who afterwards continued to contribute to BSD) nor did it stop unix vendors basing their os partly or in whole on BSD.
In fact most people would argue that BSD invogorated the unix market at the time and helped it by introducing vital new features (such as tcp/ip networking)
capitalism which encourages “the enrichment of a few at the expense of the working classes”
No, I think capitalism means that the individual can own property, in Communism everything is the State’s, and the the state is the collective of the people (in theory anyway). The individual is NOT permitted to own property.
Dictionary.com definition of Capitalism – an economic system based on private ownership of capital.
For Amiga users there are 2 boards and 3 operating systems to choose from.
And none of them are viable. I used BeOS because it made my (at the time) fast hardware run faster. As nice as the Amiga environment might be, none of it makes up for a G4 running at < 1GHz. For cripe’s sake — the Amiga was famous for it’s multimedia prowess. What kind of cutting-edge multimedia can you do on an old G4???
The Sinclair QL “died” in the early 80’s but you can get variations today.
Again, let’s compare their price/performance ratio to a modern Apple or PC machine.
You can get RiscOS machines from multiple manufacturers.
And all of them run something like a 600MHz ARM chip. No thanks!
If you want BeOS you can still get it, but it’s called Zeta now.
Zeta is DOA. I’m sorry, but if a company doesn’t have the source for the product they are hawking, then their product is stillborn.
Haiku will take a while yet but it’ll be available sometime and that’s only 2 of the choices.
I’m not terribly optimistic about Haiku either. Remember, BeOS was the media OS. These days, you can’t have a media OS that doesn’t have hardware OpenGL. A lot of the traits that a media OS should have (world-class VM, robust I/O subsystem with support for advanced multiple disk management, support for large memories arranged in a NUMA architecture, etc) will take a huge amount of time to develop on Haiku, if they ever develop at all.
I’m not terribly optimistic about Haiku either. Remember, BeOS was the media OS. These days, you can’t have a media OS that doesn’t have hardware OpenGL. A lot of the traits that a media OS should have (world-class VM, robust I/O subsystem with support for advanced multiple disk management, support for large memories arranged in a NUMA architecture, etc) will take a huge amount of time to develop on Haiku, if they ever develop at all.
Yep, and that’s why I continue to believe that B.E.OS has the right idea, but since Haiku seems to have the BeOS community mindshare and those developers call linux a “server kernel” things don’t look that bright.
Right on about Zeta. Without the code its like a dying man grasping for his last breath.
I know that hitler was against communists (and the rest of the world) as much as communism was against hitler (and the rest of the world). (the fact that by history it happened that they went face to face against one another is)
The reason for what I posted regarding hitler and communism was satiric and to show the stupidity of all these general ___ is ___ type statements that the trolls spew
“Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer)…
A bigger problem is the fact that once a user buys an OS, he is locked into the OS and possibly the hardware. This is something that is practically unique to the software industry.”
This is utter nonsense. While making a CD with a new OS is dirt cheap, the software on that CD isn’t. It represents years of investment of hundreds of programmers, testers and writers. I have absolutely no problem with IBM, Sun, etc selling their software, so long as they adhere to open standards and work towards interoperability with other vendors.
As regards lock-in, think again. This is true on some systems, like Windows, but not so much on Unix. If it were, Linux wouldn’t be eating into the Unix market! Hardware lock-in isn’t always an issue either – many open-source OSes run on Unix hardware.
I find it sad that Bryan once mentioned GPL in a rather large blog entry and that is all that is being discussed in this forum.
There is a lot more meat in there that is certainly worthy of discussion. He has some real concerns in there. The comments attached to his blog entry have been more informative and useful than what I have seen so far here.
Surely there are folks who read OSNews who are concerned about the same issues?
Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer), so the company gets considerably more then its money back per copy. In the software economy what costs money to the company is not something that is per product, like say in the car/most other industries. Of course in the car industry you are paying for the reaserch and design that went into the car as it is piggybacked onto the price of the car, but, compared to the price of the actual car, it is not that big. Simmilar things exsist in the Music/Video/Book(?) industry, but there the piggybacking of the non-per-copy cost is also relativly small.
A geez. So you’re saying that the cost is very low for a company to develop, say, an OS? Sure, the manufacturing cost is cheap but that still matters little as the development costs are so big compared to the manufacturing costs (virtually nil). People cost money. Work is not free. Get that into your head. Sure I’d like a world where we all could just not care and do whatever we please but someone have to make sure that there’s food on the table. If it costs so little to develop software then why is everyone moving their software development to India?
Tyr (IP: —.kabel.telenet.be)
History has already proven you wrong. The BSD people (original BSD Unix) gave away their os on tape with all source. This didn’t stop companies hiring those BSD hackers (who afterwards continued to contribute to BSD) nor did it stop unix vendors basing their os partly or in whole on BSD.
In fact most people would argue that BSD invogorated the unix market at the time and helped it by introducing vital new features (such as tcp/ip networking)
But the original poster talked about GPL, not BSD. The BSD license allow a company to develop a closed product whereas GPL forces the company to release the sourcecode to their product. There is a big difference in the licenses in this and I’m not gonna argue which I think is the better one (been there, done that before on OSNews).
By Anonymous (IP: 81.196.246.—)
Why is WinBlows on every desktop, NOT Solaris ???????!!!!!!
Linux will replace both !!!!!!!!
You sir/madam, are a dimwit of the largest kind (apology for the insult). I would never ever want to replace two OSes with a third. We want diversity and not monoculture. Monoculture is badness of the worst kind as it stifles development and evolution. BTW, why Windows and not Solaris? To put simply, Windows was cheap and sufficient compared to Solaris at the time the desktop war raged.
And OT about Stallman and the stupid comments about communism/Hitler/Stalin (yes, even if made as a joke, they are stupid). While I dislike Stallman and his no-compromise attitude, I find it very disturbing that you would even put his name in the same context as Hitler and Stalin. Hitler and Stalin was massmurderers of a kind that have few parallells in our time and should not be used as likness in this way.
We want diversity and not monoculture. Monoculture is badness of the worst kind as it stifles development and evolution. BTW, why Windows and not Solaris? To put simply, Windows was cheap and sufficient compared to Solaris at the time the desktop war raged.
If You don’t want monoculture why You use Windows ?
Windowd is Monoculture.
Windows == thousands of Viruses,Trojans,Spyware.
How many Viruses,trojans, spywares are written for Linux ?
Very few.
RedHat make a lot fo money from RH Advanced Server
and is based on GPL code.
Montavista, BlueCat Linux involved a lot of research
in Real Time Operating Systems ,thinks can’t be done by “bunch of hackers over internet”.
RTOS can involve
Robotics,
Automotive,
Microelectronics
High End Graphics
and Finnaly Linux && GLP Code && A lot of research in many fields. And a lot of money.
Maybe the only chance for small firms to make a profit.
Proprietary software an very expensive patents kill small firms and projects witl low budget.
First of all, I really shouldn’t reply to your comment since your chosen subject says it all, but I’ll have to point out a few misconceptions in your post.
If You don’t want monoculture why You use Windows ?
Windowd is Monoculture.
First. Who said I use only Windows? I use Windows as one of the platforms I use for computing and they involve OpenBSD, FreeBSD, Slackware Linux and Solaris (and occasionally OS X when I have to fix problems on them).
I use the tool that is best for the task at hand and that involves using different OSes.
Second. Did I say that todays situation was good with the Windows dominance? I said that I do not want to see that Windows marketshare being replaced by a Linux one, that’s was all I said.
Windows == thousands of Viruses,Trojans,Spyware.
How many Viruses, trojans, spywares are written for Linux?
You just showed the “beauty” of monoculture.
This can be stopped if people were more “educated” on computer usage.
RedHat make a lot fo money from RH Advanced Server
and is based on GPL code.
Mainly through the need for companies to have support.
and Finnaly Linux && GLP Code && A lot of research in many fields. And a lot of money.
Maybe the only chance for small firms to make a profit
Not sure exactly what your point here was. Research costs quite a lot of money. How can small firms do this without some way of being sure to get ROI as GPL (if used) lets anyone take their code into their software? Company A does research and then makes software A. They are forced (by libs etc) to make their sourcecode available through GPL. Company B see their sourcecode and makes some minor modifications to it and produce software B. Because company B did not have the research costs that A had it will be able to sell app B much cheaper than company A is. Is this good?
Proprietary software an very expensive patents kill small firms and projects witl low budget.
Proprietary software and patents are two different beasts. There is no problem with having a patent for something and also having it available as FOSS. What you cannot do is use that sourcecode unless you get the patentholders permission (usually by forking out doe). This is why software patents are bad (I never said they were good).
Proprietary software is “owned” (you can buy a copy) by a company but there is nothing that prevents you from reimplementing the feature in another app. This is reverse engineering (happends all the time).
Mix patents and proprietary software and you get a mixture for monopoly which is baaad.
I ask you now? How can a small company make a profit with GPL unless they sell support?
It’s sad to see that research in OS is actually dead. There’s no reason to do any more fancy things because the payback is not worth it because no sooner do you announce something new, you get a bunch of 12year olds screaming is it GPL’ed?.
To tell the truth, I don’t think this is all that recent of a development. With some exceptions, academic CS has diverged quite markedly from commercial CS. In particular, precious little of the academic OS research done in the last 15 years has found it’s way into production systems, while very few academic projects have been focused on the sort of problems currently at the forefront of commercial concerns.
In most engineering fields, there is a sort of cycle wherein academic research (including research-oriented divisons of commercial companies) leads commercial implementation by 5-10 years. However, that doesn’t seem to be the case in the computer industry. The current state-of-the-art in the commercial industry is a good 20 years behind academia. I think the inherent inertia of the computer industry may have a lot to do with this.
For example:
– When Lockheed makes an advancement in engine design, the cost of bringing that advancement to market is bounded by the cost of bringing it to production-quality.
– When IBM makes an advancement in computers, the cost of bringing it to market is unbounded. Not only does it have to be brought to a production level of quality, but they have to spend a tremendous amount of money in clearing out the existing product in that segment. This cost can dwarf the cost of actually developing the new product.
The bottom-line is that it’s simply not profitable for research CS and commercial CS to coexist any longer. The research folks have no interest in rehashing existing products, and the commercial folks cannot afford to shake things up too much. Thus, each goes their own way, serving their own needs.
if a bunch of 12year olds screaming is it GPL’ed? is the only thing thats holding you back from doing OS research then i guess you never was really thinking about doing it.
that would be like saying nope im not going to school cus some kids make noise during class i think il just stay home and sleep.
if you cant deal with such small issues then you have a problem and serisule need to deal with it
It’s interesting to note that there is a key component of the academia/commercial cycle that is drastically scaled-down in the computer industry. Commercial used to run highly research-oriented divisions dedicated to developing new, risky, potentially profitable ideas. These groups were key in mediating between sources of new ideas, and practical implementations of them.
These days, most of those have been drastically cut back. Apple’s Cambridge labs are closed. DEC is out of business. A lot of small, innovative companies have gone out of business. Xerox’s PARC and HP Labs have been doing a fraction of the work they used to. Microsoft spends a lot on research, but none of the cool stuff ever seems to show up in it’s products. IBM and Intel still do some good work, but they can’t carry the industry by themselves.
“if a bunch of 12year olds screaming is it GPL’ed? is the only thing thats holding you back from doing OS research then i guess you never was really thinking about doing it. ”
LOL! SO TRUE MAN! It’s so plain annoying screaming is it GPL’d? if its not GPL’d it will fail. You can’t beat the GPL. Mother, are you GPL’d? only good mothers and fathers are GPL’d.
There is a good reason why people ask “is it GPL.” The market has shown that small innovative companies have a very marginal chance of surviving. GPL (or some OSS license) gives a upstart technology the chance to survive independent of it’s company. As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
There is a good reason why people ask “is it GPL.” The market has shown that small innovative companies have a very marginal chance of surviving. GPL (or some OSS license) gives a upstart technology the chance to survive independent of it’s company. As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
Maybe it’s better to ask if it’s open-sourced instead of just GPL’d. The problem with BeOS is that none of the code was ever released after Palm bought it – except for some leaked bits I guess. If you read why Haiku is under a MIT/X11 license(I think), it’s because a major contributor gave a significant chunk of code to the project under the stipulation that it not be GPL’d. The GPL is just too viral for many people.
If you read why Haiku is under a MIT/X11 license(I think), it’s because a major contributor gave a significant chunk of code to the project under the stipulation that it not be GPL’d.
I’m curious, who was that contributor?
@Lumbergh: OSS, sure. I’m not pedantic. As long as the code doesn’t die with the company, I’m happy.
@Wolf: I’d appreciate it if you could post the research that led you to that conclusion. I’ve tried Googling for it, but all I turn up is uninformed speculation…
here is what i have to say…
What is the income of all the linux vendors combined together? So how many employees can they hire? OSS specially GPL’ed software is killing UNIX business, SUN fired so many employees…did anyone hire the people who lost jobs at SUN?
GPL is now harming windows too and thousands of other companies make money off their products on windows…but they can’t do same on linux because linux folks are used to getting stuff for free. So in a way GPL is affecting the whole software industry ECOSYSTEM.
In a healthy ecosystem, universities will do research, industry will use their research and hire those research professional, market the product, make money off it…donate some money back to universities. It was more like a self feeding system.
With GPL in picture, the whole ecosystem is broken, somebody do research, marks their software GPL, companies now can’t hire them, unless they are thinking of making money off support (which you know aint much, because support is not a specialization), so due to GPL, industry is not benefited, these researcher can’t get a job, their product is not market’ed and thus total collapse of ecosystem. GPL is like a self eating and feeding system. Its Viral, infectious.
How much innovation is done in Linux world as compared to other commerical companies like SUN, Microsoft and IBM. Sure IBM can use OSS to make money off hardware but not every other company can do it.
Hence…i believe GPL is killing software industry. Richard Stallman is Hitler.
it is one possibility that GPLing, or the demand that you do is killing the OS and software market. Another possibility is that we are consuming all our coding energys in protecting our cash flow stream. If it were all GPLed then we would get over figuring out how keep our cash flow from Office or our OS or what ever. How much energy is wasted because one thing is not open source, so we have to go and make another thing just like it to make it work.
If we got over the idea of making our billons off of something here already today… our energy today is consumed damming up the stream for our interests. We should be widening opening up the stream so it flows fast and freely for all,,, sorta like the right to read.,,, that would free up the energy that is needed to make the next great project.
What you consider an ecosystem, I consider an outmoded hierarchy. People are free, and thus they are free to code as they please, without having to adhere to some feudalistic food chain. Any argument against the GPL is silly at face-value, because they argue that people acting freely and magnanimously is somehow a bad thing. In a free society, such arguments are unacceptable.
Open source came about because some people wanted things that commercial companies were not willing to provide. So they built these things themselves. Now, if commercial companies feel cheated, well, they dug their own graves. If the market is using Linux instead of buying your products, well, maybe you should think about making your product more attractive to the market. And if you cannot make your product more appealing (free is hard to argue with), then you simply just accept that the people do not want what you have to sell. Either way, the customer wins — he get’s a better product, or a cheaper product. And in a free-market society, what’s good for the customer is ultimately good for everyone.
You made an interesting statement:
In a healthy ecosystem, universities will do research, industry will use their research
One of the prime problems is that the industry is *not* using their research. Longhorn will mark Microsoft’s transition to peddling 1980’s technology, a step-up from the 1970’s technology it was peddling before. If industry had been following academia, well, then maybe some people would want to buy their products. For example: I’ve been musing about the shortcomings of IDEs (after heavily using KDevelop and VS.NET for the last few months). If IBM hadn’t given up it’s research on Smalltalk, and had spent the last 15 years working on a killer IDE for that, they’d have a product I’d shell out $2000 for in a blink. Instead, they gave that up, and are just now (with Eclipse) catching up to where they were in the early 1990’s.
Hence…i believe GPL is killing software industry. Richard Stallman is Hitler.
I call Godwin’s law! This discussion is over, nothing to see, Wolf killed it. Great job!
Hence…i believe GPL is killing software industry. Richard Stallman is Hitler.
He is so correct! After all, since OSS (or is it FLOSS) is based on communistic ideals, then Stallman is Hitler! All of this ties into the terrorists that run Linux.
How dare people demand that the industry remain dynamic and mutate as the world changes? I think we should all ban cars as they killed the train industry (on of the most powerful and monopolistic industries back in the day)
Also all those fancy assembly lines caused all those poor workers to get fired. We want manual labor back!
And how the internet has practically transformed enourmus areas of the economy (banking, ebay, etc…).
Let as not embrace advancements if it should, heaven forbid, force the economy to remain dynamic and change to fit the needs of the time.
</sarcasm>
Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer), so the company gets considerably more then its money back per copy. In the software economy what costs money to the company is not something that is per product, like say in the car/most other industries. Of course in the car industry you are paying for the reaserch and design that went into the car as it is piggybacked onto the price of the car, but, compared to the price of the actual car, it is not that big. Simmilar things exsist in the Music/Video/Book(?) industry, but there the piggybacking of the non-per-copy cost is also relativly small.
A bigger problem is the fact that once a user buys an OS, he is locked into the OS and possibly the hardware. This is something that is practically unique to the software industry.
As a BeOS user, I’ve been burned before by a great product that didn’t stick around. From now on, when I am choosing a prospective platform, you better bet I’ll ask “Is it GPL (or OSS)?” Otherwise, I have no reason to believe the product won’t stop development tomorrow.
This is a myth. How many platforms have actually died?
If a system has users it usually continues in some form often with multiple choices:
For Amiga users there are 2 boards and 3 operating systems to choose from.
The Sinclair QL “died” in the early 80’s but you can get variations today.
You can get RiscOS machines from multiple manufacturers.
If you want BeOS you can still get it, but it’s called Zeta now. Haiku will take a while yet but it’ll be available sometime and that’s only 2 of the choices.
What does usually seem to happen is a pause, sometimes lasting years but many platforms have continued and ended up vastly more advanced than the original. If anything the “death” of a system may be one of the best things that can happen to it!
Geniuses like Wolf here can be naturally-born, I guess, but they’re often products of machinery that has interests in their unusual intelligence and is also willing to support it financially.
I’m here mostly for the comments and I sure wouldn’t be that democratic, if it was my site.
He is so correct! After all, since OSS (or is it FLOSS) is based on communistic ideals, then Stallman is Hitler!
I’m going to post even though Godwin’s Law has been called. 😉 Just a history correction. Nazi’s were strongly anti-communist. If OSS is based on communistic ideals, Stallman must be Joseph Stalin.
The existance of FLOSS is not the consequence of Stallman’s ideas, it is related to the very nature of software which can be copied at zero cost. There is no alternative. Software can be compared with Music and Litterature whose artworks are not based on physical artefacts ( unlike painting or theatre … )
CS research can be made at a very low cost, by individuals or in third world countries, it is basically a cheap science, comparing with Physics, Chemistry, Biology, … ( It is nearly as cheap as Maths ). Software patents is a mean to transform zero investment ideas into unfair concurrency barriers.
Commercial software vendors have abandoned for long the idea of catching up with CS research. I think that they fear progress. For example, in Lisp Machine or Smalltalk style environments, where every single bit of applications is available for reuse or tuning, it is impossible to sell something like a closed source word processor. Microsoft, Adobe, and others cannot follow CS research because really advanced computer environments are a threat for their existance ( Imagine an oil company doing research for propelling cars with water instead of petroleum ).
Why FLOSS developpers doesn’t try to make these ideas widely available still puzzles me.
( BTW, really advanced computers doesn’t need programmers, they will do programming themselves 🙂 <grin/>
( About Hitler & Communism : He was strongly anti-everyone. He arranged an anti-aggression pact with Stalin at the begginning of WW2, while prosecuting german communists. Communism is not about making everything free to everyone, it’s about nationalisation of production means, as an opposition to capitalism which encourages “the enrichment of a few at the expense of the working classes”. In other words, a communist US gov. would nationalize Microsoft and sell Windows $2 to everyone. IMHO, Stallman is more an Anarchist than a Communist. )
As I once said in osnews (and IIRC was promptly modded), the danger I see in having software (or anything) for free (or perceived as such) is stifle in innovation. Tthis has nothing to do with MS pr – MS isn’t innovative by any means, and the fact that its OS is ‘perceived as free’ – in the sense that when Joe user buys a computer it comes with it – is the greatest contributor to the mentioned stifling effect.
That happens because, when there is a tool for a job that you can have for nothing, you tend to stick with it, no matter its shortcomings. Just as the availability of slave labour discourages technological research (but especially adoption), so does the availability of ‘reasonable’ software.
In a market where you have to pay for products, you tend to pick what suits you best. In a monopoly (as MS) or a market where products are considered to come for free (the GPL), you are at a loss when it comes to decide what’s best. Both stifle competition. Now the latter is much preferable to the former as the individual user is concerned, but in large scale I’m not convinced the difference is big.
It would be nice if the ‘software industry’ were like most others: companies would ask a _fair_ price for their software, be open to customer requests, and struggle to make their products better! That is, the opposite of what went on in the 80s and 90s mainstream.
I’m going to post even though Godwin’s Law has been called. 😉 Just a history correction. Nazi’s were strongly anti-communist. If OSS is based on communistic ideals, Stallman must be Joseph Stalin.
Interesting enough, I once had an US right winger argue that nationalsocialism was in fact a “leftist” thing, much like communism. The proof? Well, it started with the name. Nationalsocialism. Some people…
GPL is now harming windows too and thousands of other companies make money off their products on windows…but they can’t do same on linux because linux folks are used to getting stuff for free. So in a way GPL is affecting the whole software industry ECOSYSTEM.
History has already proven you wrong. The BSD people (original BSD Unix) gave away their os on tape with all source. This didn’t stop companies hiring those BSD hackers (who afterwards continued to contribute to BSD) nor did it stop unix vendors basing their os partly or in whole on BSD.
In fact most people would argue that BSD invogorated the unix market at the time and helped it by introducing vital new features (such as tcp/ip networking)
capitalism which encourages “the enrichment of a few at the expense of the working classes”
No, I think capitalism means that the individual can own property, in Communism everything is the State’s, and the the state is the collective of the people (in theory anyway). The individual is NOT permitted to own property.
Dictionary.com definition of Capitalism – an economic system based on private ownership of capital.
For Amiga users there are 2 boards and 3 operating systems to choose from.
And none of them are viable. I used BeOS because it made my (at the time) fast hardware run faster. As nice as the Amiga environment might be, none of it makes up for a G4 running at < 1GHz. For cripe’s sake — the Amiga was famous for it’s multimedia prowess. What kind of cutting-edge multimedia can you do on an old G4???
The Sinclair QL “died” in the early 80’s but you can get variations today.
Again, let’s compare their price/performance ratio to a modern Apple or PC machine.
You can get RiscOS machines from multiple manufacturers.
And all of them run something like a 600MHz ARM chip. No thanks!
If you want BeOS you can still get it, but it’s called Zeta now.
Zeta is DOA. I’m sorry, but if a company doesn’t have the source for the product they are hawking, then their product is stillborn.
Haiku will take a while yet but it’ll be available sometime and that’s only 2 of the choices.
I’m not terribly optimistic about Haiku either. Remember, BeOS was the media OS. These days, you can’t have a media OS that doesn’t have hardware OpenGL. A lot of the traits that a media OS should have (world-class VM, robust I/O subsystem with support for advanced multiple disk management, support for large memories arranged in a NUMA architecture, etc) will take a huge amount of time to develop on Haiku, if they ever develop at all.
I’m not terribly optimistic about Haiku either. Remember, BeOS was the media OS. These days, you can’t have a media OS that doesn’t have hardware OpenGL. A lot of the traits that a media OS should have (world-class VM, robust I/O subsystem with support for advanced multiple disk management, support for large memories arranged in a NUMA architecture, etc) will take a huge amount of time to develop on Haiku, if they ever develop at all.
Yep, and that’s why I continue to believe that B.E.OS has the right idea, but since Haiku seems to have the BeOS community mindshare and those developers call linux a “server kernel” things don’t look that bright.
Right on about Zeta. Without the code its like a dying man grasping for his last breath.
I know that hitler was against communists (and the rest of the world) as much as communism was against hitler (and the rest of the world). (the fact that by history it happened that they went face to face against one another is)
The reason for what I posted regarding hitler and communism was satiric and to show the stupidity of all these general ___ is ___ type statements that the trolls spew
“Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer)…
A bigger problem is the fact that once a user buys an OS, he is locked into the OS and possibly the hardware. This is something that is practically unique to the software industry.”
This is utter nonsense. While making a CD with a new OS is dirt cheap, the software on that CD isn’t. It represents years of investment of hundreds of programmers, testers and writers. I have absolutely no problem with IBM, Sun, etc selling their software, so long as they adhere to open standards and work towards interoperability with other vendors.
As regards lock-in, think again. This is true on some systems, like Windows, but not so much on Unix. If it were, Linux wouldn’t be eating into the Unix market! Hardware lock-in isn’t always an issue either – many open-source OSes run on Unix hardware.
I find it sad that Bryan once mentioned GPL in a rather large blog entry and that is all that is being discussed in this forum.
There is a lot more meat in there that is certainly worthy of discussion. He has some real concerns in there. The comments attached to his blog entry have been more informative and useful than what I have seen so far here.
Surely there are folks who read OSNews who are concerned about the same issues?
By Anonymous (IP: —.sothfd01.mi.comcast.net)
Serioussly though, the whole software economy is completly broken. You pay hundreds of dollars for a copy of a product. Creating that copy was free (as in beer), so the company gets considerably more then its money back per copy. In the software economy what costs money to the company is not something that is per product, like say in the car/most other industries. Of course in the car industry you are paying for the reaserch and design that went into the car as it is piggybacked onto the price of the car, but, compared to the price of the actual car, it is not that big. Simmilar things exsist in the Music/Video/Book(?) industry, but there the piggybacking of the non-per-copy cost is also relativly small.
A geez. So you’re saying that the cost is very low for a company to develop, say, an OS? Sure, the manufacturing cost is cheap but that still matters little as the development costs are so big compared to the manufacturing costs (virtually nil). People cost money. Work is not free. Get that into your head. Sure I’d like a world where we all could just not care and do whatever we please but someone have to make sure that there’s food on the table. If it costs so little to develop software then why is everyone moving their software development to India?
Tyr (IP: —.kabel.telenet.be)
History has already proven you wrong. The BSD people (original BSD Unix) gave away their os on tape with all source. This didn’t stop companies hiring those BSD hackers (who afterwards continued to contribute to BSD) nor did it stop unix vendors basing their os partly or in whole on BSD.
In fact most people would argue that BSD invogorated the unix market at the time and helped it by introducing vital new features (such as tcp/ip networking)
But the original poster talked about GPL, not BSD. The BSD license allow a company to develop a closed product whereas GPL forces the company to release the sourcecode to their product. There is a big difference in the licenses in this and I’m not gonna argue which I think is the better one (been there, done that before on OSNews).
By Anonymous (IP: 81.196.246.—)
Why is WinBlows on every desktop, NOT Solaris ???????!!!!!!
Linux will replace both !!!!!!!!
You sir/madam, are a dimwit of the largest kind (apology for the insult). I would never ever want to replace two OSes with a third. We want diversity and not monoculture. Monoculture is badness of the worst kind as it stifles development and evolution. BTW, why Windows and not Solaris? To put simply, Windows was cheap and sufficient compared to Solaris at the time the desktop war raged.
And OT about Stallman and the stupid comments about communism/Hitler/Stalin (yes, even if made as a joke, they are stupid). While I dislike Stallman and his no-compromise attitude, I find it very disturbing that you would even put his name in the same context as Hitler and Stalin. Hitler and Stalin was massmurderers of a kind that have few parallells in our time and should not be used as likness in this way.
We want diversity and not monoculture. Monoculture is badness of the worst kind as it stifles development and evolution. BTW, why Windows and not Solaris? To put simply, Windows was cheap and sufficient compared to Solaris at the time the desktop war raged.
If You don’t want monoculture why You use Windows ?
Windowd is Monoculture.
Windows == thousands of Viruses,Trojans,Spyware.
How many Viruses,trojans, spywares are written for Linux ?
Very few.
RedHat make a lot fo money from RH Advanced Server
and is based on GPL code.
Montavista, BlueCat Linux involved a lot of research
in Real Time Operating Systems ,thinks can’t be done by “bunch of hackers over internet”.
RTOS can involve
Robotics,
Automotive,
Microelectronics
High End Graphics
and Finnaly Linux && GLP Code && A lot of research in many fields. And a lot of money.
Maybe the only chance for small firms to make a profit.
Proprietary software an very expensive patents kill small firms and projects witl low budget.
GPL is a chance for them.
First of all, I really shouldn’t reply to your comment since your chosen subject says it all, but I’ll have to point out a few misconceptions in your post.
If You don’t want monoculture why You use Windows ?
Windowd is Monoculture.
First. Who said I use only Windows? I use Windows as one of the platforms I use for computing and they involve OpenBSD, FreeBSD, Slackware Linux and Solaris (and occasionally OS X when I have to fix problems on them).
I use the tool that is best for the task at hand and that involves using different OSes.
Second. Did I say that todays situation was good with the Windows dominance? I said that I do not want to see that Windows marketshare being replaced by a Linux one, that’s was all I said.
Windows == thousands of Viruses,Trojans,Spyware.
How many Viruses, trojans, spywares are written for Linux?
You just showed the “beauty” of monoculture.
This can be stopped if people were more “educated” on computer usage.
RedHat make a lot fo money from RH Advanced Server
and is based on GPL code.
Mainly through the need for companies to have support.
and Finnaly Linux && GLP Code && A lot of research in many fields. And a lot of money.
Maybe the only chance for small firms to make a profit
Not sure exactly what your point here was. Research costs quite a lot of money. How can small firms do this without some way of being sure to get ROI as GPL (if used) lets anyone take their code into their software? Company A does research and then makes software A. They are forced (by libs etc) to make their sourcecode available through GPL. Company B see their sourcecode and makes some minor modifications to it and produce software B. Because company B did not have the research costs that A had it will be able to sell app B much cheaper than company A is. Is this good?
Proprietary software an very expensive patents kill small firms and projects witl low budget.
Proprietary software and patents are two different beasts. There is no problem with having a patent for something and also having it available as FOSS. What you cannot do is use that sourcecode unless you get the patentholders permission (usually by forking out doe). This is why software patents are bad (I never said they were good).
Proprietary software is “owned” (you can buy a copy) by a company but there is nothing that prevents you from reimplementing the feature in another app. This is reverse engineering (happends all the time).
Mix patents and proprietary software and you get a mixture for monopoly which is baaad.
I ask you now? How can a small company make a profit with GPL unless they sell support?